Hi John,
Thanks for the response. These were created by Analyze 6.0 from 130
single slice dicom images from a Philips Intera. I'm away from my image
machine just now, but can send a hdr if it will help.
I thought if the origin field was empty, that SPM assumed the mid-point.
Cheers,
Mark
John Ashburner wrote:
>Dear Mark,
>Where did you get them, and how did you convert them to a format that SPM can
>(try to) read? They have probably been converted using some software that
>puts information in the same place in the header as the ORIGIN field used by
>SPM. One possibility is that they are NIFTI-1 (which SPM2 doesn't recognise
>- but SPM5 will).
>
>Best regards,
>-John
>
>
>
>>I have some (less than ideal) MRI structural images that I want to
>>co-register with PET data. When I display them in SPM2 it is reporting
>>the following:
>>
>>Dimensions: 512x512x130 (correct)
>>Datatype: int16 (correct)
>>Intensity: Y=1X (correct)
>>
>>Vox size: -0.449 x 0.449 x 1.39 (correct)
>>Origin: 1.8e+04 1.77e+04 1.72e+04 (strange)
>>Dir Cos: 1.000 0.000 0.000
>> 0.000 0.540 -0.841
>> 0.000 0.841 0.540
>>
>>The images were acquired approximately coronally, so I re-orient by a
>>ptich rotation of approx -pi/2 before I did this display.
>>If I click on the image and get the crossharis to about were I would
>>like the origin to be the display window reports the crosshair position as:
>>
>>mm: 7973.1 15786.8 -19443.1
>>vx: 253.2 309.1 60.0
>>
>>It seems to me that SPM2 is not handling the voxel to world translation
>>correctly. Can anyone help?
>>
>>Many thanks,
>>
>>Mark
>>
>>
|